Hello everybody,
I did a differnce-in-difference anlysis for in health care expenditures using proc genmod and %NLEstimate. People participating in a stuctured health care programme (Prog=1) were compared to those who do not participate (Prog=0). With data for both groups availabe for 1 year before intervention (post=0) and 1 year after intervention (post=1). The same subjects are measured before and after. The code I used works for the primary outcome "total cost". But now I need to analyse subcategories of costs, additinally (for hospitalisation, drugs,...) and some of these subcategories are zero-inflated.
Is there any option how I can make a DiD anaysis using zero-inflated data?
Any help would be greatly appreciated.
Maggie
proc genmod data= cost;
class prog post / ref= first;
model total_cost= prog post prog*post / dist=gamma link=log;
estimate "diff_in_diff" prog*post 1 -1 -1 1;
lsmeans prog*post / e linik;
lsestimate prog*post "diff_in_diff" 1 -1 -1 1;
store out= gammod;
run;
data difdif;
input k1-k4;
set=1;
datalines;
1 -1 -1 1;
%NLEstimate (instore= gammod,
label= Diff in Diff
f=(exp(b_p1+b_p2+b_p4+b_p6) - exp(b_p1-b_p4)) -
(exp(b_p1+b_p2) - exp(b_p1)),
title= DiD Totel cost)
